- Email: [email protected]
Capecitabine and oxaliplatin in advanced colorectal cancer: A dose-finding study
Annals of Oncology 12 1737-1741.2001. © 2001 Kluwer Academic Publishers. Printed in the Netherlands.
Original article Capecitabine and oxaliplatin in advanced colorectal cancer: A dose-finding study M. Zeuli,1 F. Di Costanzo,2 A. Sdrobolini,2 S. Gasperoni,2 F.P. Paoloni,1 A. Carpi,3 L. Moscetti,4 R. Cherubini5 & F. Cognetti1 on behalf of Gruppo Oncologico Italiano per la Ricerca Clinica (GOIRC) and Gruppo Oncologico Laziale (GOL) 1
Department of Medical Oncology A, Regina Elena Cancer Institute, Rome; Department of Medical Oncology, Azienda Ospedaliera S Maria, Terni, ' Department of Medical Oncology, Arcispedale S. Maria Nuova, Reggio Emilia; *Department of Medical Oncology. Universita La Sapienza', Rome; 5 Department of Medical Oncology, Azienda Ospedaliera, Perugia, Italy
Summary Purpose; Capecitabine and oxaliplatin are both active anticancer agents in the treatment of patients with advanced colorectal cancer (ACRC). The aim of this dose-finding trial was to determine the maximum-tolerated dose (MTD), the doselimiting toxicities (DLTs) and the activity of the combination in patients with advanced colorectal cancer. Patients and methods; Twenty-five chemotherapy-pretreated patients received the combination of capecitabine and oxaliplatin. Capecitabine was administered orally twice a day continuously for 14 days in doses ranging from 1650 to 2500 mg/m 2 /d, and oxaliplatin was administered as a two-hour infusion on day 1 using dose, ranges from 100 to 130 mg/m 2 repeated every three weeks. Results; Twenty-five patients were assessable for toxicity,
Introduction
In the last few years, several new oral fluoropyrimidines (Doxifluridina, UFT, 5-etynyl-uracil, S-l and capecitabine) have been developed in an attempt to maintain or improve the effectiveness of intravenous 5-fluorouracil (5-FU). The potential advantages of oral administration include a associated cost reduction, a better patient compliance and a reduction in toxic effects (neutropenia and stomatitis). These new fluoropyrimidines provide prolonged 5-FU exposure at lower peak concentration than continuous infusion (CI) of 5-FU. Capecitabine (Xeloda, Roche Pharmaceuticals, Basel, Switzerland) was developed as an orally administered pro-drug of 5-FU and tumor selective citotoxic agent [1, 2]. After oral administration, capecitabine passes rapidly unchanged through the gastrointestinal mucosal barrier as the intact molecule and is extensively absorbed with low variability. In a phase I trial, capecitabine was administered on a continuous schedule or intermittent schedule (two weeks followed by a week's rest) with or without folinic acid
and DLTs were diarrhea (grade ^ 3: 27%) and stomatitis (grade > 3 : 9%) at dose level VI. Dose level V (capecitabine 2500 mg/m 2 and oxaliplatin 120 mg/m 2 ) was found to be the MTD. Hematological toxicity was minimal, overall neurotoxicity (grade 1-4) was 27% with 1% grade 3-4. A global response rate was 17% (95% confidence interval (95% Cl): 2%-32%) and the median overall survival was 12 months. Conclusion: The recommended dose for further phase II studies is capecitabine 2500 mg/m 2 /d with intermittent schedule and oxaliplatin 120 mg/m 2 every three weeks. The toxicities were mainly gastrointestinal: diarrhea, stomatitis and vomiting. This combination should be studied in phase II trials in advanced colorectal. Key words: capecitabine, colorectal cancer, dose-finding, oxaliplatin
(FA) [2]. Recommended dose and schedule of capecitabine is 2500 mg/m" without FA and the most adverse events (AE) and dose limiting toxicities (DLT) were hand-foot syndrome (HFS), nausea/vomiting, diarrhea, fatigue, anorexia and stomatitis. Remarkable antitumor activity was found with intermittent regimen in patients with colorectal cancer and breast cancer which were heavily pretreated with regimens including 5-FU. Two randomized trials were started in USA and Europe plus Australasia involving patients with advanced colorectal cancer (ACRC) previously untreated by chemotherapy (CT) [3,4]. Both these trials compared capecitabine administered according to an intermittent schedule vs. the Mayo Clinic schedule. Both trials showed a better response rate (RR) with capecitabine (27%-23% vs. 18%-15%) and demonstrated no difference in progression free-survival (PFS) and overall survival (OS), although toxicity grade 3-4 (diarrhea and neutropenia) was higher in the 5-FU plus FA schedule. Oxaliplatin, a new third generation cisplatin analogue in the 1,2-diaminocyclohexane family of platinum compounds, has recently been developed for clinical use
1738 Table I Dose escalation and DLTs. Dose level
Capecitabine (nig/nT) Oxaliplatin (mg/nr) No. ofpts No. of cycles Median Range DLT No DLT type
1
II
Ill
IV
V
VI
1650 100 3 12 3 3-6 -
2000 100 3 12 3 3-6 -
2000 110 3 15 6 3-6 -
2500 110 3 12 3 1-8 -
2500 120 10 33 2
2500 130 3 11 3 3-5
[5]. The main limiting toxicity of oxaliplatin is a sensory neuropathy (paresthesia and dysesthesia) in the extremities and perioral area, triggered or enhanced by exposure to cold. Oxaliplatin was first developed as a combination therapy with 5-FU plus FA administered as a chronomodulate infusion over five days [6]. In a phase II trial the RR was 18% in previously untreated patients with an OS of approximately 13-14 months and 10% in patients refractory to 5-FU with OS of 8-10 months [7]. In combination with 5-FU and FA in randomized phase III trials, it showed a twofold or greater increase in RR and a three-month gain in time to progression (TTP), with survival difference blurred by a crossover effect [8, 9]. We report the results of a multicentric dose-finding study of the combination of capecitabine and oxaliplatin in patients with ACRC previously treated with 5-FU to determine the maximum-tolerated dose (MTD) and to elicit a recommended dose for further clinical development.
Patients and methods Patients with histologically-proven advanced adenocarcinotna of the colon or rectum were eligible for this study They had to be pretreated with no more than two CT lines for advanced disease or relapsed within six months after the end of adjuvant 5-FU-based adjuvant therapy. Measurable disease was not required. All patients were required to have World Health Organization (WHO) performance status (PS) ^ 2 . age between 18 and 75 years, life expectancy greater than 12 weeks, WBC count 4 x 109 cells/1, absolute neutrophil count (ANC) of 2.0 x 109 neutrophils/l, platelet count 100 x 109 platelets/1, hemoglobin level 10 g/dl, serum creatinine level within normal values, total bilirubin level 1.5 times the upper limit of normal (ULN). alkaline phosphatases less than five times ULN. AST and ALT 1.5 times ULN (or < 5 x ULN if hepatic metastases were present). Exclusion criteria were: CTor radiotherapy within four weeks before study entry, prior CT with capecilabine or oxaliplatin, history of other malignancies (except excised in situ cervical carcinoma or basal/squamous cell skin carcinoma), peripheral neuropathy of National Cancer Institute (NCI) grade 2, symptomatic brain metastases and concurrent treatment with other experimental drugs or participation in another clinical trial within 30 days prior to study screening Written informed consent was required for each patient.
1-8 1 Diarrhea
?
Diarrhea/ stomatitis
Pretreatment evaluation included a complete clinical history, physical examination, neurological status. ECG. chest X-ray, and computed tomography scan or ultrasound of assessable target lesions. Complete blood cell counts and differential blood chemistry were obtained at baseline and were repeated weekly. Toxicities were graded using the NCI common toxicity criteria DLTs were defined as grade 4 leuko/neutropenia or grade 4 thrombocytopenia or any grade ^ 3 non hematologic toxicity except grade 3 nausea and/or vomiting. Objective response was a secondary end point and was recorded according to standard WHO response criteria [10], The planned dose escalation of both drugs is listed in Table 1. Three patients at each dose level were recruited. Before escalating to the next dose level, at least three patients should have received one cycle of chemotherapy and been monitored for any acute toxicity during a first cycle (three-week observation period). If one of three patients at one dose level developed a DLT, an additional three patients were accrued at the same dose level. If an additional DLT was observed at this level, no further escalation was allowed and the previous dose level was declared MTD. Once the MTD was defined, three more patients were treated at this dose level. No intrapatient dose escalation was allowed. Treatment was continued until progressive disease (PD) or for a maximum of one year. Capecitabine was supplied as film-coated tablets in two dose strengths (150 mg and 500 mg) and was delivered as a daily treatment for two weeks followed by one week of rest. It was self-administered at patients' homes in the form of two divided doses approximately 12 hours apart, taken within 30 minutes after ingestion of food. Oxaliplatin (Sanofi-Synthelabo, Paris, France) was administered as a two-hour infusion on day one of the treatment. This treatment cycle was repeated every 21 days. Before chemotherapy, patients were systematically given intravenous ondansetron or granisetron plus steroids to prevent emesis. Patients were warned to avoid exposure to cold and cold drinks after therapy with oxaliplatin to reduce related neuropathy episodes or their exacerbation. In the case of grade 2 neurotoxicity, oxaliplatin infusion duration was further lengthened to four or six hours if the patients had neurotoxicity superior to grade 2.
Results Patient characteristics Twenty-five patients were enrolled and all the patients were treated on an outpatient basis. A total of 95 cycles of chemotherapy were delivered with a median of five cycles per patient (range 1-9). No patients died from study-related events. Patient characteristics are listed in Table 2. Fifteen patients (60%) had a PS of 0 and 10 (40%) of 1. Eight patients had received adjuvant CT
1739 according to the Majo Clinic regimen. Thirteen patients were pretreated for advanced disease with modulation of 5-FU (bolus) by FA (47%) or MTX (9%), eight patients pretreated with the DeGramont's schedule alone (13%) or plus Irinotecan (21%) and two (9%) patients had received Raltitrexed or Irinotecan alone. Two untreated patients were enrolled because they had relapsed within six months after adjuvant therapy. The median number of previous CT lines was 1 (range 0-2). All patients had liver metastases. Ten patients (40%) had only one site of disease, five (20%) had two sites and 10 (40%) had three or more. Determination of the MTD Table 1 summarizes the frequency of the DLTs that were considered for MTD. From dose level I to the first cohort of patients enrolled at dose level V, no DLTs were observed, so dose escalation to the VI level was pursued. At dose level VI (capecitabine 2500 mg/m 2 /d and oxaliplatin 130 mg/m 2 ), two of three patients had DLTs (grade 4 diarrhea and grade 4 stomatitis) requiring dose reduction after the second cycle, and this level was considered the toxic dose. Three further patients were enrolled at level V (capecitabine 2500 mg/m 2 /d and oxaliplatin 120 mg/m 2 ) and DLT (grade 4 diarrhea) was observed in 1 of 6, so in order to better assess the safety profile of the combination at this dose level, we enrolled four further patients in which no DLTs were observed. According to our results, dose level V was deemed to be the recommended dose for phase II trials, despite a high proportion of gastrointestinal toxicity grade ^ 2, including severe diarrhea in one out often patients. Safety
Table 2. Patient characteristics. Total patients
25
Median age (range) Sex Male/female (%) ECOG PS 0/1 (%) Sites of disease Liver Lung Nodes Bone Other No. of sites of disease 1 2 »3 Previous CT" Adjuvant Advanced disease Mayo FOLFIRI de Gramont Irinotecan MTX/FU Raltitrexed Median no of previous CT lines (range)
66 (43-75) 14/11 (56/44) 15/10 (60/40) 25 (100) 5(20) 5(20) 2(8) 10 (40) 10 (40) 5(20) 10 (40) 8(35) 23 (92) II 5 3 1 2 1 1 (0-2)
Mayo:
FA 20 mg/nr/d 1° -• 5° q 28 days 5-FU 375 mg/m 2 /d 1° - 5° q 28 days Irinotecan 180 mg/m 2 /d (90-min inf) 1°. q 14 days FOLFIRI FA 200 mg/nr/d (2-hr inf) 1°. 2° q 14 days 5-FU 400 mg/m 2 /d (bolus inf) 1°. 2° q 14 days 5-FU 400 mg/nr/d (22-hr inf) 1 °. 2° q 14 days 2 de Gramont: FA 200 mg/m /d (2-hr inf) 1°. 2° q 14 days 5-FU 400 mg/nr/d (bolus inf) 1°. 2° q 14 days 5-FU 400 mg/nr/d (22-hr inf) I °. 2° q 14 days 2 Irinotecan: Irinotecan 350 mg/m /d (90-min inf) 1°, q 21 days Methotrexate 200 mg/m 2 /d (30-min inf) 1° q 14 days MTX/FU. 5-FU 600 mg/nr/d (bolus inf) 2° q 14 days FA 10 mg/m 2 every 6 hr x 6 (p.o.) 3° q 14 days Ratritrexed: Raltitrexed 3 mg/nr/d (15-min inf) 1°, q 21 days
During the first cycle no toxicity greater than grade 2 was observed in the cohorts of patients from dose levels I-IV: nausea and vomiting were the most common adverse events and occurred in 6 of 12 patients; but, although frequent, such events were mild and rapidly as well as fully reversible (grade 1 or 2). Grade 2 diarrhea occurred only in 3 of 12 patients. This event occurred at sequent cycles in one (grade 4). One patient experienced dose level III in two patients and at dose level IV in one. grade 4 stomatitis at dose level VI, leading to the disOther gastrointestinal events observed included mild continuation of treatment. A transient and asymptomatic stomatitis (one patient) and transient epigastralgias increase in liver transaminases was observed in one (one patient) at dose levels III and IV, respectively. patient. Hematologic toxicities were neither frequent Grade 1 peripheral neurotoxicity was observed in only nor severe, and no hand-foot syndrome was observed one patient at dose level I. in the first treatment cycle at all dose levels. Five out often patients treated at dose level V experiTable 3 lists non-hematological toxicities overall. Tolenced diarrhea, reaching grade 4 criteria in one. At dose erance was excellent from dose level I-IV, with only one level V, grade 1 nausea or vomiting was observed in 4 of main toxicity in 50 cycles administered: one patient 10 patients whereas, grade 2 vomiting was recorded in alone had grade 3 peripheral neurotoxicity after three only one patient. Severe (grade 3) vomiting occurred in treatment cycles at dose level IV. Nausea and neuroall patients treated at the highest dose. At dose level VI, logical toxicity were the most common adverse events diarrhea was also frequent and severe, occurring in two and occurred in 28% and 27% of the courses of treatof three patients and requiring dose reduction in sub- ment, respectively; but while frequent, the effect was
1740 Table 3 Non-hematologic study-related adverse events, any grade, per cycle, per patient and per dose level. Dose level
No of cycles (pts)
Nausea grade 1-2 /3-4
Vomiting grade 1-2 / 3 - 4
Diarrhea grade 1-2 / 3-4
Stomatitis grade 1-2 / 3-4
Neurotoxicity grade 1-2 / 3-4
Asthenia
HFS
1 II III IV V VI Total
12(3) 12(3) 15(3) 12(3) 33(10) 11(3) 95(25)
3 (3) / 5 (3) / 6 (3) / 6 (3) / 6(5)/1(0/27(18) / -
5 (2) / 2(2)/4 (3) / 1(1)/3 (3) / - / 3 (3) 16 ( 1 2 ) / 3 (3)
- /2(1)/4(2)/1(1)/6 (5)/2(1) 3 ( 2 ) / 1(1) 16 ( 1 1 ) / 3 (2)
-/-/1(1)/1(1)/-/-/ I d ) 2 ( 2 ) / 1(1)
1(1)/2(1)/3 (2) / 6 ( 2 ) / 1(1) 6 (3) / 6(2)/2 5 ( 1 1 ) / 1(1)
1(1) 4(2) 4(2) 1(1) 10(6)
7(1) 7(1)
generally mild (grade 1 or 2). Grade 3 vomiting occurred in 8%, 20%, 9% and 36% of courses of treatment at dose level I, III, Vand VI, respectively. Diarrhea was the third most frequent non-hematological toxicity, occurring in 19% of the courses of treatment. Its incidence was also dose-dependent, two cases of grade 4 diarrhea being observed only at the two highest dose levels. Stomatitis was observed in only 3% of cycles, but one patient at VI dose level experienced a grade 4 stomatitis leading to the discontinuation of treatment. Mild HFS was seen in only one patient treated at dose level IV. Two patients experienced transient laringospasm during the first oxaliplatin infusion. Hematologic toxicity was moderate. Anemia was the most frequent hematological toxicity occurring in seven patients (29% of cycles). Grade 3 neutropenia occurred only in one patient treated at dose level V. No febrile neutropenia was reported. No patients had thrombocytopenia grade 3-4. Efficacy Response assessment was not the primary end point of this study. We observed four objective responses (OR) in 23 assessable patients (17%). Two patients were not evaluable for efficacy: one refused treatment after the first cycle and one was withheld due to a concomitant uncontrolled non chemotherapy-related infection. One patient pretreated with two lines of chemotherapy (Mayo regimen and irinotecan) achieved a complete response (CR) on multiple liver metastases with a progression free survival (PFS) of 18 weeks, and three patients had a partial clinical response (PR) lasting 14, 17 and 24 weeks. They were all pretreated with only one line of chemotherapy for advanced disease (Mayo regimen). In five patients, stable disease (SD) was observed, and fourteen patients had progressive disease (PD). The median survival was 12 months (range 1-20), and after 12 months 53% of patients are alive.
Discussion 5-FU remains one of the most important drugs in the treatment of ACRC. Several studies have confirmed that
Laringospasm
_ 2(2) 2(2)
CI of 5-FU is superior to bolus intravenous injection, the former resulting in improved anticancer activity as well as reduced toxicity. However, CI requires an indwelling central venous catheter with associated increased risks of infection and thrombosis. The CI increases the costs of this palliative treatment and probably the quality of life of patients. Therefore, the development of effective oral fluoropirimidine is a very desiderable goal in ACRC. In this study, we assessed the DLTand the MTD of the capecitabine/oxaliplatin combination in patients pretreated for ACRC. No unexpected adverse event occurred, and the most common toxicities observed were those of each drug, except for peripheral neuropathy, observed in only 16% of patients, and HFS (in a procedure that prevents 4%). This could be due to the low median number of cycles administered per patient, the development of cumulative toxicities. Mild gastrointestinal toxicities were seen in approximately 80% of patients. Main toxicities occurred at dose level V (capecitabine 2500 mg/m 2 and oxaliplatin 120 mg/m 2 ) and VI (capecitabine 2500 mg/m 2 and oxaliplatin 130 mg/m 2 ) and consisted of vomiting diarrhea and stomatitis, all of which seem to be the DLTs of this regimen. No severe neutropenia or thrombocytopenia occurred. It is noteworthy that the recommended dose of each drug in this combination regimen is as high as that recommended for single-agent therapy. At this dose level, only one patient out of ten in our study experienced grade 3 toxicity. The capecitabine/oxaliplatin combination seems less myelosuppressive than the combination of oxaliplatin with 5-FU flat CI [12]. Severe diarrhea occurred as frequently with oxaliplatin combined with capecitabine than with bolus, flat CI or chronomodulated 5-FU [11-14]. Severe vomiting seems to have occurred more frequently than with bolus or flat CI of 5-FU. Severe stomatitis occurred rarely with the capecitabine/oxaliplatin combination. Hence, it is possible that capecitabine, when combined with oxaliplatin, exhibits the lowest incidence of severe toxicities of all 5-FU regimens. Four patients (17%) experienced OR with a median duration of 17.5 weeks, and five (21%) patients had SD. This result is interesting because many patients were heavily pretreated in our study. Thirty-five
1741 percent were pretreated with CI of 5-FU plus FA. All responding patients were pretreated with bolus 5-FU, while no patient previously treated with CI 5-FU responded. This finding supports the hypothesis that capecitabine simulates CI 5-FU, achieving responses in patients pretreated with bolus regimens but has no activity in Cl-refractory ones. A good safety profile and a promising antitumoral activity was therefore seen in a European phase I study in which patients received a fixed dose of oxaliplatin 130 mg/m 2 combined with one of four doses of capecitabine (1000, 1650, 2000 and 2500 mg/nr/d) administered on an intermittent schedule. The recommended phase II trial dose was reached at dose level III (capecitabine 2000 mg/m 2 /d) and five out of nine patients with ACRC had a PR [12]. In conclusion, capecitabine/oxaliplatin combination seems to be a safe regimen and the ease of administration of this schedule makes it acceptable in the palliative treatment of patients with ACRC. A phase II trial is ongoing in our group to evaluate the activity in untreated patients.
7.
8
9.
10. 11.
12.
13.
References 1. Twelves C, Budman DR. Creaven PJ et al. Pharmacokinetics (PK.) and Pharmacodynamics (PD) of capecitabine in two phase I studies. Proc Am Soc Clin Oncol 1996; 15: 476 (Abstr 1509). 2. Dirix LY, Bisset D, Van Oosterom AT et al. A phase I study of capecitabine in combination with oral leucovorin in patients with advanced and/or metastatic solid cancer. Proc Am Soc Clin Oncol 1996; 15. 496 (Abstr 1821). 3. Cox JV, Pazdur R.Thibault A et al. A phase III trial of Xeloda™ (capecitabine) in previously untreated advanced/metastatic colorectal cancer. Proc Am Soc Clin Oncol 1999; (Abstr 1016) 4. Twelves C, Harper P, Van Cutsem E et al. A phase III trial of Xeloda™ (capecitabine) in previously untreated advanced/metastatic colorectal cancer. Proc Am Soc Clin Oncol 1999: (Abstr 1010). 5. Burchenal J, Kalaker K, Dew K et al. Rationale for development of platinum analogs. Cancer Treat Rep 1979; 63: 1493-7. 6. Garufi C, Brienza S, Bensmaine MA et al. Addition of oxaliplatin
14.
(L-OHP) to chronomodulated (CM) 5-fluorouracil and folinic acid for reversal of acquired chemoresistance in patients with advanced colorectal cancer (ACC). Proc Am Soc Clin Oncol 1995; 14: 192. Levi F. Zidani R. Vannetzel JM et al. Chronomodulated vs. fixed infusion rate delivery of ambulatory chemotherapy with oxaliplatin. 5-fluorouracil and folinic acid in patients with colorectal cancer metastases: A randomized multi-institutional trial. J Natl Cancer Inst 1994; 86: 1608-17. Giacchetti S, Perpoint B. Zigani R et al. Phase 111 multicentcr randomized trial of oxaliplatin added to chronomodulated fluorouracil-leucovorin as first-line treatment of metastatic colorectal cancer. J Clin Oncol 2000; 18: 136-47. De Gramont A. Figer A. Seymour M et al. Leucovorin and fluorouracil with or without oxaliplatin as first-line treatment in advanced colorectal cancer. J Clin Oncol 2000; 18: 2938-47. Miller AB, Hoogstraten B. Staquet M et al. Reporting results of cancer treatment. Cancer 1981; 47: 207-14. Giacchetti S, Itzahaki M. Adam R et al. Long term survival of patients with unresectable colorectal cancer liver metastases. following infusional chronotherapy with 5-fluorouracil (5-FU). folinic acid (FA), oxaliplatin (L-OHP). Ann Oncol 1996. 7 (Suppl 5): 34. Diaz-Rubio E, Evans J. Tabernero J et al. Phase 1 study of capecitabine in combination with oxaliplatin in patients with advanced or metastatic solid tumors. Proc Am Soc Clin Oncol 2000 (Abstr 198a). Zon Comba A. Blajman C. Richardet E et al. Bimonthly oxaliplatin (L-OHP) with (A) or without (B) fluorouracil (FU) and leucovorin (FA): Proven evidence of synergism in a phase II randomized trial. Proc Am Soc Clin Oncol 1999: 18: 248a (Abstr 953). Levi F, Zidani R, Misset JL et al International organization for cancer chronotherapy: Randomised multicentre trial of chronotherapy with oxaliplatin. fluorouracil and folinic acid in metastatic colorectal cancer. Lancet 1997; 350: 681-6.
Received 21 May 2001; accepted 20 July 2001.
Correspondence to • M. Zeuh, MD Oncologia Medica A Istituto Regina Elena - Polo Oncologico Via Elio Chianesi n 53 00144 Rome Italy E-mail: massimozeuli@tiscalinet it
Original article Capecitabine and oxaliplatin in advanced colorectal cancer: A dose-finding study M. Zeuli,1 F. Di Costanzo,2 A. Sdrobolini,2 S. Gasperoni,2 F.P. Paoloni,1 A. Carpi,3 L. Moscetti,4 R. Cherubini5 & F. Cognetti1 on behalf of Gruppo Oncologico Italiano per la Ricerca Clinica (GOIRC) and Gruppo Oncologico Laziale (GOL) 1
Department of Medical Oncology A, Regina Elena Cancer Institute, Rome; Department of Medical Oncology, Azienda Ospedaliera S Maria, Terni, ' Department of Medical Oncology, Arcispedale S. Maria Nuova, Reggio Emilia; *Department of Medical Oncology. Universita La Sapienza', Rome; 5 Department of Medical Oncology, Azienda Ospedaliera, Perugia, Italy
Summary Purpose; Capecitabine and oxaliplatin are both active anticancer agents in the treatment of patients with advanced colorectal cancer (ACRC). The aim of this dose-finding trial was to determine the maximum-tolerated dose (MTD), the doselimiting toxicities (DLTs) and the activity of the combination in patients with advanced colorectal cancer. Patients and methods; Twenty-five chemotherapy-pretreated patients received the combination of capecitabine and oxaliplatin. Capecitabine was administered orally twice a day continuously for 14 days in doses ranging from 1650 to 2500 mg/m 2 /d, and oxaliplatin was administered as a two-hour infusion on day 1 using dose, ranges from 100 to 130 mg/m 2 repeated every three weeks. Results; Twenty-five patients were assessable for toxicity,
Introduction
In the last few years, several new oral fluoropyrimidines (Doxifluridina, UFT, 5-etynyl-uracil, S-l and capecitabine) have been developed in an attempt to maintain or improve the effectiveness of intravenous 5-fluorouracil (5-FU). The potential advantages of oral administration include a associated cost reduction, a better patient compliance and a reduction in toxic effects (neutropenia and stomatitis). These new fluoropyrimidines provide prolonged 5-FU exposure at lower peak concentration than continuous infusion (CI) of 5-FU. Capecitabine (Xeloda, Roche Pharmaceuticals, Basel, Switzerland) was developed as an orally administered pro-drug of 5-FU and tumor selective citotoxic agent [1, 2]. After oral administration, capecitabine passes rapidly unchanged through the gastrointestinal mucosal barrier as the intact molecule and is extensively absorbed with low variability. In a phase I trial, capecitabine was administered on a continuous schedule or intermittent schedule (two weeks followed by a week's rest) with or without folinic acid
and DLTs were diarrhea (grade ^ 3: 27%) and stomatitis (grade > 3 : 9%) at dose level VI. Dose level V (capecitabine 2500 mg/m 2 and oxaliplatin 120 mg/m 2 ) was found to be the MTD. Hematological toxicity was minimal, overall neurotoxicity (grade 1-4) was 27% with 1% grade 3-4. A global response rate was 17% (95% confidence interval (95% Cl): 2%-32%) and the median overall survival was 12 months. Conclusion: The recommended dose for further phase II studies is capecitabine 2500 mg/m 2 /d with intermittent schedule and oxaliplatin 120 mg/m 2 every three weeks. The toxicities were mainly gastrointestinal: diarrhea, stomatitis and vomiting. This combination should be studied in phase II trials in advanced colorectal. Key words: capecitabine, colorectal cancer, dose-finding, oxaliplatin
(FA) [2]. Recommended dose and schedule of capecitabine is 2500 mg/m" without FA and the most adverse events (AE) and dose limiting toxicities (DLT) were hand-foot syndrome (HFS), nausea/vomiting, diarrhea, fatigue, anorexia and stomatitis. Remarkable antitumor activity was found with intermittent regimen in patients with colorectal cancer and breast cancer which were heavily pretreated with regimens including 5-FU. Two randomized trials were started in USA and Europe plus Australasia involving patients with advanced colorectal cancer (ACRC) previously untreated by chemotherapy (CT) [3,4]. Both these trials compared capecitabine administered according to an intermittent schedule vs. the Mayo Clinic schedule. Both trials showed a better response rate (RR) with capecitabine (27%-23% vs. 18%-15%) and demonstrated no difference in progression free-survival (PFS) and overall survival (OS), although toxicity grade 3-4 (diarrhea and neutropenia) was higher in the 5-FU plus FA schedule. Oxaliplatin, a new third generation cisplatin analogue in the 1,2-diaminocyclohexane family of platinum compounds, has recently been developed for clinical use
1738 Table I Dose escalation and DLTs. Dose level
Capecitabine (nig/nT) Oxaliplatin (mg/nr) No. ofpts No. of cycles Median Range DLT No DLT type
1
II
Ill
IV
V
VI
1650 100 3 12 3 3-6 -
2000 100 3 12 3 3-6 -
2000 110 3 15 6 3-6 -
2500 110 3 12 3 1-8 -
2500 120 10 33 2
2500 130 3 11 3 3-5
[5]. The main limiting toxicity of oxaliplatin is a sensory neuropathy (paresthesia and dysesthesia) in the extremities and perioral area, triggered or enhanced by exposure to cold. Oxaliplatin was first developed as a combination therapy with 5-FU plus FA administered as a chronomodulate infusion over five days [6]. In a phase II trial the RR was 18% in previously untreated patients with an OS of approximately 13-14 months and 10% in patients refractory to 5-FU with OS of 8-10 months [7]. In combination with 5-FU and FA in randomized phase III trials, it showed a twofold or greater increase in RR and a three-month gain in time to progression (TTP), with survival difference blurred by a crossover effect [8, 9]. We report the results of a multicentric dose-finding study of the combination of capecitabine and oxaliplatin in patients with ACRC previously treated with 5-FU to determine the maximum-tolerated dose (MTD) and to elicit a recommended dose for further clinical development.
Patients and methods Patients with histologically-proven advanced adenocarcinotna of the colon or rectum were eligible for this study They had to be pretreated with no more than two CT lines for advanced disease or relapsed within six months after the end of adjuvant 5-FU-based adjuvant therapy. Measurable disease was not required. All patients were required to have World Health Organization (WHO) performance status (PS) ^ 2 . age between 18 and 75 years, life expectancy greater than 12 weeks, WBC count 4 x 109 cells/1, absolute neutrophil count (ANC) of 2.0 x 109 neutrophils/l, platelet count 100 x 109 platelets/1, hemoglobin level 10 g/dl, serum creatinine level within normal values, total bilirubin level 1.5 times the upper limit of normal (ULN). alkaline phosphatases less than five times ULN. AST and ALT 1.5 times ULN (or < 5 x ULN if hepatic metastases were present). Exclusion criteria were: CTor radiotherapy within four weeks before study entry, prior CT with capecilabine or oxaliplatin, history of other malignancies (except excised in situ cervical carcinoma or basal/squamous cell skin carcinoma), peripheral neuropathy of National Cancer Institute (NCI) grade 2, symptomatic brain metastases and concurrent treatment with other experimental drugs or participation in another clinical trial within 30 days prior to study screening Written informed consent was required for each patient.
1-8 1 Diarrhea
?
Diarrhea/ stomatitis
Pretreatment evaluation included a complete clinical history, physical examination, neurological status. ECG. chest X-ray, and computed tomography scan or ultrasound of assessable target lesions. Complete blood cell counts and differential blood chemistry were obtained at baseline and were repeated weekly. Toxicities were graded using the NCI common toxicity criteria DLTs were defined as grade 4 leuko/neutropenia or grade 4 thrombocytopenia or any grade ^ 3 non hematologic toxicity except grade 3 nausea and/or vomiting. Objective response was a secondary end point and was recorded according to standard WHO response criteria [10], The planned dose escalation of both drugs is listed in Table 1. Three patients at each dose level were recruited. Before escalating to the next dose level, at least three patients should have received one cycle of chemotherapy and been monitored for any acute toxicity during a first cycle (three-week observation period). If one of three patients at one dose level developed a DLT, an additional three patients were accrued at the same dose level. If an additional DLT was observed at this level, no further escalation was allowed and the previous dose level was declared MTD. Once the MTD was defined, three more patients were treated at this dose level. No intrapatient dose escalation was allowed. Treatment was continued until progressive disease (PD) or for a maximum of one year. Capecitabine was supplied as film-coated tablets in two dose strengths (150 mg and 500 mg) and was delivered as a daily treatment for two weeks followed by one week of rest. It was self-administered at patients' homes in the form of two divided doses approximately 12 hours apart, taken within 30 minutes after ingestion of food. Oxaliplatin (Sanofi-Synthelabo, Paris, France) was administered as a two-hour infusion on day one of the treatment. This treatment cycle was repeated every 21 days. Before chemotherapy, patients were systematically given intravenous ondansetron or granisetron plus steroids to prevent emesis. Patients were warned to avoid exposure to cold and cold drinks after therapy with oxaliplatin to reduce related neuropathy episodes or their exacerbation. In the case of grade 2 neurotoxicity, oxaliplatin infusion duration was further lengthened to four or six hours if the patients had neurotoxicity superior to grade 2.
Results Patient characteristics Twenty-five patients were enrolled and all the patients were treated on an outpatient basis. A total of 95 cycles of chemotherapy were delivered with a median of five cycles per patient (range 1-9). No patients died from study-related events. Patient characteristics are listed in Table 2. Fifteen patients (60%) had a PS of 0 and 10 (40%) of 1. Eight patients had received adjuvant CT
1739 according to the Majo Clinic regimen. Thirteen patients were pretreated for advanced disease with modulation of 5-FU (bolus) by FA (47%) or MTX (9%), eight patients pretreated with the DeGramont's schedule alone (13%) or plus Irinotecan (21%) and two (9%) patients had received Raltitrexed or Irinotecan alone. Two untreated patients were enrolled because they had relapsed within six months after adjuvant therapy. The median number of previous CT lines was 1 (range 0-2). All patients had liver metastases. Ten patients (40%) had only one site of disease, five (20%) had two sites and 10 (40%) had three or more. Determination of the MTD Table 1 summarizes the frequency of the DLTs that were considered for MTD. From dose level I to the first cohort of patients enrolled at dose level V, no DLTs were observed, so dose escalation to the VI level was pursued. At dose level VI (capecitabine 2500 mg/m 2 /d and oxaliplatin 130 mg/m 2 ), two of three patients had DLTs (grade 4 diarrhea and grade 4 stomatitis) requiring dose reduction after the second cycle, and this level was considered the toxic dose. Three further patients were enrolled at level V (capecitabine 2500 mg/m 2 /d and oxaliplatin 120 mg/m 2 ) and DLT (grade 4 diarrhea) was observed in 1 of 6, so in order to better assess the safety profile of the combination at this dose level, we enrolled four further patients in which no DLTs were observed. According to our results, dose level V was deemed to be the recommended dose for phase II trials, despite a high proportion of gastrointestinal toxicity grade ^ 2, including severe diarrhea in one out often patients. Safety
Table 2. Patient characteristics. Total patients
25
Median age (range) Sex Male/female (%) ECOG PS 0/1 (%) Sites of disease Liver Lung Nodes Bone Other No. of sites of disease 1 2 »3 Previous CT" Adjuvant Advanced disease Mayo FOLFIRI de Gramont Irinotecan MTX/FU Raltitrexed Median no of previous CT lines (range)
66 (43-75) 14/11 (56/44) 15/10 (60/40) 25 (100) 5(20) 5(20) 2(8) 10 (40) 10 (40) 5(20) 10 (40) 8(35) 23 (92) II 5 3 1 2 1 1 (0-2)
Mayo:
FA 20 mg/nr/d 1° -• 5° q 28 days 5-FU 375 mg/m 2 /d 1° - 5° q 28 days Irinotecan 180 mg/m 2 /d (90-min inf) 1°. q 14 days FOLFIRI FA 200 mg/nr/d (2-hr inf) 1°. 2° q 14 days 5-FU 400 mg/m 2 /d (bolus inf) 1°. 2° q 14 days 5-FU 400 mg/nr/d (22-hr inf) 1 °. 2° q 14 days 2 de Gramont: FA 200 mg/m /d (2-hr inf) 1°. 2° q 14 days 5-FU 400 mg/nr/d (bolus inf) 1°. 2° q 14 days 5-FU 400 mg/nr/d (22-hr inf) I °. 2° q 14 days 2 Irinotecan: Irinotecan 350 mg/m /d (90-min inf) 1°, q 21 days Methotrexate 200 mg/m 2 /d (30-min inf) 1° q 14 days MTX/FU. 5-FU 600 mg/nr/d (bolus inf) 2° q 14 days FA 10 mg/m 2 every 6 hr x 6 (p.o.) 3° q 14 days Ratritrexed: Raltitrexed 3 mg/nr/d (15-min inf) 1°, q 21 days
During the first cycle no toxicity greater than grade 2 was observed in the cohorts of patients from dose levels I-IV: nausea and vomiting were the most common adverse events and occurred in 6 of 12 patients; but, although frequent, such events were mild and rapidly as well as fully reversible (grade 1 or 2). Grade 2 diarrhea occurred only in 3 of 12 patients. This event occurred at sequent cycles in one (grade 4). One patient experienced dose level III in two patients and at dose level IV in one. grade 4 stomatitis at dose level VI, leading to the disOther gastrointestinal events observed included mild continuation of treatment. A transient and asymptomatic stomatitis (one patient) and transient epigastralgias increase in liver transaminases was observed in one (one patient) at dose levels III and IV, respectively. patient. Hematologic toxicities were neither frequent Grade 1 peripheral neurotoxicity was observed in only nor severe, and no hand-foot syndrome was observed one patient at dose level I. in the first treatment cycle at all dose levels. Five out often patients treated at dose level V experiTable 3 lists non-hematological toxicities overall. Tolenced diarrhea, reaching grade 4 criteria in one. At dose erance was excellent from dose level I-IV, with only one level V, grade 1 nausea or vomiting was observed in 4 of main toxicity in 50 cycles administered: one patient 10 patients whereas, grade 2 vomiting was recorded in alone had grade 3 peripheral neurotoxicity after three only one patient. Severe (grade 3) vomiting occurred in treatment cycles at dose level IV. Nausea and neuroall patients treated at the highest dose. At dose level VI, logical toxicity were the most common adverse events diarrhea was also frequent and severe, occurring in two and occurred in 28% and 27% of the courses of treatof three patients and requiring dose reduction in sub- ment, respectively; but while frequent, the effect was
1740 Table 3 Non-hematologic study-related adverse events, any grade, per cycle, per patient and per dose level. Dose level
No of cycles (pts)
Nausea grade 1-2 /3-4
Vomiting grade 1-2 / 3 - 4
Diarrhea grade 1-2 / 3-4
Stomatitis grade 1-2 / 3-4
Neurotoxicity grade 1-2 / 3-4
Asthenia
HFS
1 II III IV V VI Total
12(3) 12(3) 15(3) 12(3) 33(10) 11(3) 95(25)
3 (3) / 5 (3) / 6 (3) / 6 (3) / 6(5)/1(0/27(18) / -
5 (2) / 2(2)/4 (3) / 1(1)/3 (3) / - / 3 (3) 16 ( 1 2 ) / 3 (3)
- /2(1)/4(2)/1(1)/6 (5)/2(1) 3 ( 2 ) / 1(1) 16 ( 1 1 ) / 3 (2)
-/-/1(1)/1(1)/-/-/ I d ) 2 ( 2 ) / 1(1)
1(1)/2(1)/3 (2) / 6 ( 2 ) / 1(1) 6 (3) / 6(2)/2 5 ( 1 1 ) / 1(1)
1(1) 4(2) 4(2) 1(1) 10(6)
7(1) 7(1)
generally mild (grade 1 or 2). Grade 3 vomiting occurred in 8%, 20%, 9% and 36% of courses of treatment at dose level I, III, Vand VI, respectively. Diarrhea was the third most frequent non-hematological toxicity, occurring in 19% of the courses of treatment. Its incidence was also dose-dependent, two cases of grade 4 diarrhea being observed only at the two highest dose levels. Stomatitis was observed in only 3% of cycles, but one patient at VI dose level experienced a grade 4 stomatitis leading to the discontinuation of treatment. Mild HFS was seen in only one patient treated at dose level IV. Two patients experienced transient laringospasm during the first oxaliplatin infusion. Hematologic toxicity was moderate. Anemia was the most frequent hematological toxicity occurring in seven patients (29% of cycles). Grade 3 neutropenia occurred only in one patient treated at dose level V. No febrile neutropenia was reported. No patients had thrombocytopenia grade 3-4. Efficacy Response assessment was not the primary end point of this study. We observed four objective responses (OR) in 23 assessable patients (17%). Two patients were not evaluable for efficacy: one refused treatment after the first cycle and one was withheld due to a concomitant uncontrolled non chemotherapy-related infection. One patient pretreated with two lines of chemotherapy (Mayo regimen and irinotecan) achieved a complete response (CR) on multiple liver metastases with a progression free survival (PFS) of 18 weeks, and three patients had a partial clinical response (PR) lasting 14, 17 and 24 weeks. They were all pretreated with only one line of chemotherapy for advanced disease (Mayo regimen). In five patients, stable disease (SD) was observed, and fourteen patients had progressive disease (PD). The median survival was 12 months (range 1-20), and after 12 months 53% of patients are alive.
Discussion 5-FU remains one of the most important drugs in the treatment of ACRC. Several studies have confirmed that
Laringospasm
_ 2(2) 2(2)
CI of 5-FU is superior to bolus intravenous injection, the former resulting in improved anticancer activity as well as reduced toxicity. However, CI requires an indwelling central venous catheter with associated increased risks of infection and thrombosis. The CI increases the costs of this palliative treatment and probably the quality of life of patients. Therefore, the development of effective oral fluoropirimidine is a very desiderable goal in ACRC. In this study, we assessed the DLTand the MTD of the capecitabine/oxaliplatin combination in patients pretreated for ACRC. No unexpected adverse event occurred, and the most common toxicities observed were those of each drug, except for peripheral neuropathy, observed in only 16% of patients, and HFS (in a procedure that prevents 4%). This could be due to the low median number of cycles administered per patient, the development of cumulative toxicities. Mild gastrointestinal toxicities were seen in approximately 80% of patients. Main toxicities occurred at dose level V (capecitabine 2500 mg/m 2 and oxaliplatin 120 mg/m 2 ) and VI (capecitabine 2500 mg/m 2 and oxaliplatin 130 mg/m 2 ) and consisted of vomiting diarrhea and stomatitis, all of which seem to be the DLTs of this regimen. No severe neutropenia or thrombocytopenia occurred. It is noteworthy that the recommended dose of each drug in this combination regimen is as high as that recommended for single-agent therapy. At this dose level, only one patient out of ten in our study experienced grade 3 toxicity. The capecitabine/oxaliplatin combination seems less myelosuppressive than the combination of oxaliplatin with 5-FU flat CI [12]. Severe diarrhea occurred as frequently with oxaliplatin combined with capecitabine than with bolus, flat CI or chronomodulated 5-FU [11-14]. Severe vomiting seems to have occurred more frequently than with bolus or flat CI of 5-FU. Severe stomatitis occurred rarely with the capecitabine/oxaliplatin combination. Hence, it is possible that capecitabine, when combined with oxaliplatin, exhibits the lowest incidence of severe toxicities of all 5-FU regimens. Four patients (17%) experienced OR with a median duration of 17.5 weeks, and five (21%) patients had SD. This result is interesting because many patients were heavily pretreated in our study. Thirty-five
1741 percent were pretreated with CI of 5-FU plus FA. All responding patients were pretreated with bolus 5-FU, while no patient previously treated with CI 5-FU responded. This finding supports the hypothesis that capecitabine simulates CI 5-FU, achieving responses in patients pretreated with bolus regimens but has no activity in Cl-refractory ones. A good safety profile and a promising antitumoral activity was therefore seen in a European phase I study in which patients received a fixed dose of oxaliplatin 130 mg/m 2 combined with one of four doses of capecitabine (1000, 1650, 2000 and 2500 mg/nr/d) administered on an intermittent schedule. The recommended phase II trial dose was reached at dose level III (capecitabine 2000 mg/m 2 /d) and five out of nine patients with ACRC had a PR [12]. In conclusion, capecitabine/oxaliplatin combination seems to be a safe regimen and the ease of administration of this schedule makes it acceptable in the palliative treatment of patients with ACRC. A phase II trial is ongoing in our group to evaluate the activity in untreated patients.
7.
8
9.
10. 11.
12.
13.
References 1. Twelves C, Budman DR. Creaven PJ et al. Pharmacokinetics (PK.) and Pharmacodynamics (PD) of capecitabine in two phase I studies. Proc Am Soc Clin Oncol 1996; 15: 476 (Abstr 1509). 2. Dirix LY, Bisset D, Van Oosterom AT et al. A phase I study of capecitabine in combination with oral leucovorin in patients with advanced and/or metastatic solid cancer. Proc Am Soc Clin Oncol 1996; 15. 496 (Abstr 1821). 3. Cox JV, Pazdur R.Thibault A et al. A phase III trial of Xeloda™ (capecitabine) in previously untreated advanced/metastatic colorectal cancer. Proc Am Soc Clin Oncol 1999; (Abstr 1016) 4. Twelves C, Harper P, Van Cutsem E et al. A phase III trial of Xeloda™ (capecitabine) in previously untreated advanced/metastatic colorectal cancer. Proc Am Soc Clin Oncol 1999: (Abstr 1010). 5. Burchenal J, Kalaker K, Dew K et al. Rationale for development of platinum analogs. Cancer Treat Rep 1979; 63: 1493-7. 6. Garufi C, Brienza S, Bensmaine MA et al. Addition of oxaliplatin
14.
(L-OHP) to chronomodulated (CM) 5-fluorouracil and folinic acid for reversal of acquired chemoresistance in patients with advanced colorectal cancer (ACC). Proc Am Soc Clin Oncol 1995; 14: 192. Levi F. Zidani R. Vannetzel JM et al. Chronomodulated vs. fixed infusion rate delivery of ambulatory chemotherapy with oxaliplatin. 5-fluorouracil and folinic acid in patients with colorectal cancer metastases: A randomized multi-institutional trial. J Natl Cancer Inst 1994; 86: 1608-17. Giacchetti S, Perpoint B. Zigani R et al. Phase 111 multicentcr randomized trial of oxaliplatin added to chronomodulated fluorouracil-leucovorin as first-line treatment of metastatic colorectal cancer. J Clin Oncol 2000; 18: 136-47. De Gramont A. Figer A. Seymour M et al. Leucovorin and fluorouracil with or without oxaliplatin as first-line treatment in advanced colorectal cancer. J Clin Oncol 2000; 18: 2938-47. Miller AB, Hoogstraten B. Staquet M et al. Reporting results of cancer treatment. Cancer 1981; 47: 207-14. Giacchetti S, Itzahaki M. Adam R et al. Long term survival of patients with unresectable colorectal cancer liver metastases. following infusional chronotherapy with 5-fluorouracil (5-FU). folinic acid (FA), oxaliplatin (L-OHP). Ann Oncol 1996. 7 (Suppl 5): 34. Diaz-Rubio E, Evans J. Tabernero J et al. Phase 1 study of capecitabine in combination with oxaliplatin in patients with advanced or metastatic solid tumors. Proc Am Soc Clin Oncol 2000 (Abstr 198a). Zon Comba A. Blajman C. Richardet E et al. Bimonthly oxaliplatin (L-OHP) with (A) or without (B) fluorouracil (FU) and leucovorin (FA): Proven evidence of synergism in a phase II randomized trial. Proc Am Soc Clin Oncol 1999: 18: 248a (Abstr 953). Levi F, Zidani R, Misset JL et al International organization for cancer chronotherapy: Randomised multicentre trial of chronotherapy with oxaliplatin. fluorouracil and folinic acid in metastatic colorectal cancer. Lancet 1997; 350: 681-6.
Received 21 May 2001; accepted 20 July 2001.
Correspondence to • M. Zeuh, MD Oncologia Medica A Istituto Regina Elena - Polo Oncologico Via Elio Chianesi n 53 00144 Rome Italy E-mail: massimozeuli@tiscalinet it